The association between Bcl-2 and cancer has been known for 20 years. But the biochemical mechanisms to explain why Bcl-2, found at translocation breakpoints in follicular lymphomas, why elevated Bcl-xL expression in many tumor types, and why herpes viruses that encode Bcl-2 homologues cause cancer, remains unknown. The field has focused considerable effort to understand the complex interactions between anti-death (Bcl-2 and Bcl-xL) and pro-death (Bax and Bak) family proteins, and their mechanisms of action after cells have received a stimulus that induced apoptosis. The work in our lab has forced us to think in another direction, to pursue the basic biochemical functions of Bcl-2 family proteins that function prior to receipt of a death stimulus, and that we predict are shared between both anti- and pro-death family members. These 'day-jobs' are also predicted to explain the anti-death functions of human Bcl-2 proteins in a remarkable diversity of species including mammals, plants and fungi, Therefore, we propose to apply the new tools available for yeast to study the 'core1 functions of Bcl-2 family proteins. First, we seek to identify genes that regulate programmed cell death in yeast. Several different death stimuli, including viruses, heat shock and metabolic stress, will be applied to the complete library of yeast knockout strains under conditions that distinguish anti- and pro-death genes. Factors that are common or unique to specific pathways will be distinguished. These pathways will be ordered and candidate yeast regulators of cell death already on hand will be further investigated to determine the connection between their ability to regulate mitochondria morphology and function. Finally, yeast will be probed to identify those genes that are required for human Bcl-2 and Bcl-xL to inhibit cell death in yeast. We expect that these newly unidentified yeast factors will be involved in the maintenance, biogenesis and degradation of mitochondria, and in close proximity to the regulation of bioenergetics. [unreadable] [unreadable] [unreadable]